‘Chamaliens’ on the Hawaiian Islands: spatial risk assessment for the invasive Jackson’s chameleon (Chamaeleonidae)

2011·2.6 MB·English

by  Rödder, D

SALAMANDRA 47(1) 36–42 20 February 2011 CoIrSrSeNsp o0n03d6e–n3ce375
Correspondence
‘Chamaliens’ on the Hawaiian Islands: spatial risk assessment
for the invasive Jackson’s chameleon (Chamaeleonidae)
Dennis Rödder 1, Wolfgang Böhme 1 & Stefan Lötters 2
1) Zoological Research Museum Alexander Koenig, Department of Herpetology, Adenauerallee 160,
53113 Bonn, Germany
2) Trier University, Department of Biogeography, Am Wissenschaftspark 25 + 27, 54296 Trier, Germany
Corresponding author: Dennis Rödder, e-mail: [email protected]
Manuscript received: 13 April 2010
Alien species are of major concern to conservation biology, al. (2009) documented predation on endemic snail and in-
agriculture and the human society, as they may become in- sect species, including some threatened with extinction,
vasive and successfully compete with native species, nega- and Hagey et al. (2010) commented on its foraging strat-
tively influence crop production, and cause health prob- egy. Further studies on the natural history and distribution
lems. Several reptile species are considered invasive alien of T. jacksonii on Hawaiian Islands, coupled with monitor-
species, but most of them are little studied (Kraus 2009). ing and management efforts, are required.
Amongst these species are a few chameleons. Apparently, GIS-based Species Distribution Models (SDMs) are a
several of the Chamaeleo chamaeleo (Linnaeus, 1758) pop- helpful tool towards these goals, as they focus on species’
ulations in the Mediterranean (e.g. Crete, Malta, Sicily) are potential distributions. In the case of invasive alien spe-
non-autochthonous and their presence can be linked to cies, they can be understood as spatial risk assessments.
historical trade routes (e.g. Klaver 1981). Chamaeleo af­ Technically, a SDM assesses a species’ ecological niche and
ricanus Laurenti, 1768 from northern Africa was translo- projects it into geographical space (Rödder et al. 2010a).
cated from the Nile delta onto the Peloponesos where it has Niche information is obtained at presence and sometimes
maintained reproducing populations perhaps since antiq- absence sites of the study species; climatic niche informa-
uity (e.g. Böhme et al. 1998, Böhme 2000). More recently, tion has been demonstrated to be a good predictor for in-
the Malagasy Furcifer pardalis (Cuvier, 1829) has become vasion success in alien reptile species (e.g. Bomford et
invasive on Réunion Island (e.g. Bourgat 1970). al. 2009; Rödder et al. 2009b). The resulting map shows
One of the most successful invasive chameleons is Jack- similarities with the niche elsewhere, e.g. by different grid
son’s chameleon, Trioceros jacksonii (Boulenger, 1896), values, interpreted as ‘likelihood’ of the species’ potential
which is native to Afromontane Kenya and Tanzania occurrence (e.g. Peterson & Vieglais 2001, Rödder &
(Fig. 1). Popular as a pet and commonly traded and im- Lötters 2010). Projecting SDMs onto past or future cli-
ported by the thousands from Kenya and Tanzania (Car- mate scenarios may allow simulations of a species’ poten-
penter et al. 2004), this species was introduced to Oahu, tial response to environmental changes (e.g. Rödder &
Hawaiian Islands in 1972 (e.g. McKeown 1991, Waring Schulte 2010). For example, Rödder (2009) provided a
1997). Apparently, established populations are referable spatial risk assessment for the invasive non-autochthonous
to the large, up to 35 cm in total length, subspecies T. j. frog Eleutherodactylus coqui on the Hawaiian Islands, sug-
xantholophus (Eason, Ferguson & Hebrard, 1988) from gesting range allocation towards higher elevations under
Kenya and Tanzania (McKeown 1991) (Fig. 2). As a result future anthropogenic climate change, thus affecting pro-
of inter-island transport which has only been restricted tected areas.
since 1997, Jackson’s chameleon currently has established In this paper, we develop SDMs for Jackson’s chameleon
wild populations on all the main islands including the is- in the Hawaiian Islands under current and future climates
land of Hawaii, Maui and Kauii as a result of multiple intro- based on the species’ climatic envelope at its invaded and
ductions and subsequent spread (Fig. 1; Waring 1997 cited native African distribution ranges. The goal is to appreci-
in Hagey et al. 2010). The species is especially common in ate the geographical potential of this species in its invaded
wetter habitats at elevations from 100–1,000 m above sea range.
level (McKeown 1996, Holland et al. 2009) and inhab- For SDM building, we used presence data from the na-
its disturbed habitats as well as forested areas (McKeown tive African and the invaded Hawaiian ranges of Trioceros
1991, Waring 1997). The ecological impact of Hawaiian T. jacksonii as available via the Global Biodiversity Informa-
jacksonii has not been studied. Only recently, Holland et tion Facility (http://www.gbif.org) and HerpNet (http://
© 2011 Deutsche Gesellschaft für Herpetologie und Terrarienkunde e.V. (DGHT), Rheinbach, Germany
A36ll articles available online at http://www.salamandra-journal.com
Correspondence
Figure 1. Species records used for model building in T. jacksonii’s invaded Hawaiian range and its native range in Kenya and Tanzania
(insert).
www.herpnet.org). When necessary, georeferencing of lo- of coordinates. Records were only considered if they could
cality data was conducted with the BioGeoMancer (http:// be unambiguously assigned to a single grid cell, leaving a
bg.berkeley.edu/latest/). DIVA-GIS 7.1.6 (Hijmans et al. total number of 81 presence records for modelling (Fig. 1).
2002; http://www.diva-gis.org) was used to check accuracy Information on current climate, i.e., for the period
1950–2000, was obtained from the WorldClim database
at a spatial resolution of 30 arc sec (Hijmans et al. 2005;
http://www.worldclim.org). For future climate, as expected
for the year 2080, we downloaded interpolations of three
different global change scenarios of the third and fourth
assessments of the IPCC (2007); spatial resolution 30 arc
sec (Ramirez & Jarvis 2008; http://gisweb.ciat.cgiar.org/
GCMPage). These anthropogenic climate change pre-
dictions, simulating conditions as expected in the two
IPCC story lines A2a and B2a, were derived from simu-
lations of three climate models CCCma-CGCM2 (Flato
et al. 2000, Flato & Boer 2001), CSIRO-MK2 (Gordon
& O´Farrell 1997), and UKMO-HadCM3 (Gordon et
al. 2000, Pope et al. 2000). Climate variables comprised
monthly minimum and maximum temperatures, and aver-
age precipitation per month. Based on these, we computed
19 ‘bioclimatic’ variables with DIVA-GIS; these are com-
monly used for species distribution modelling (e.g. Busby
1991, Beaumont et al. 2005). Since multicolinearity of pre-
dictor variables may hamper successful SDM projection
through space or time (e.g. Heikkinen et al. 2006), we
extracted all ‘bioclimatic’ values from the species records
and performed a pair-wise correlation analysis based on
Pearson’s correlation coefficient with XLSTAT 2010 (http://
www.adinsoft.com). For SDM computation, only variables
with R2 < 0.75 were considered, resulting in a final variable
set comprising ’mean monthly temperature range’, ’tem-
perature seasonality’, ’maximum temperature of the warm-
est month’, ’minimum temperature of the coldest month’, ’
annual temperature range’, ’annual precipitation’, ’precipita-
tion in the wettest month’, ’precipitation seasonality’, ’pre-
cipitation in the warmest quarter’ and ’precipitation in the
coldest quarter’. It was suggested that these variables de-
scribe environmental conditions that are physiologically
Figure 2. Non-collected male Trioceros jacksonii xantholophus important to T.  jacksonii (Bennett 2004, Andrews 2008,
from Parklands in Nairobi. Lin 1979, Lin & Nelson 1980).
37
Correspondence
Figure 3. Comparisons of the climate spaces occupied in T. jack­
sonii’s native and invaded Hawaiian ranges.
For SDM development, Maxent 3.2.2 was applied (Phil- “2009”), and potential impact of climate change (Rödder
lips et al. 2006, Phillips & Dudík 2008), a machine- & Weinsheimer 2009, Rödder et al. 2010b). In most cas-
learning algorithm following the principle of maximum es, Maxent outperforms comparable SDM methods, even
entropy (Jaynes 1957). Maxent has been successfully ap- when the sample size is small (Hernandez et al. 2006,
plied in studies assessing possible spreads of invasive al- Wisz et al. 2008). This was further confirmed by Maxent’s
ien species (Giovanelli et al. 2007, Rödder & Lötters ability to predict previously unknown populations in poor-
2010), pathogens (Rödder et al. 2009a, Lötters et al. 2010 ly known species (Pearson et al. 2007, Weinsheimer et
38
Correspondence
current
A2a
B2a
Figure 4. Potential distribution of Jackson’s chameleon on the Hawaiian Islands under current climate and two future anthropogenic
climate change scenarios (A2a, B2a), based on a Maxent Species Distribution Model and bioclimatic conditions at the species’ native
African occurrence. Warmer colours indicated higher suitability to the species when considering climate conditions only (each left)
and when considering climate as well as land cover (each right). Established reserves are indicated as crosshatched areas.
al. 2010) and even species that are likely to be new to sci- the species records, equalling the target group background
ence (Raxworthy et al. 2003). Next to species presence approach suggested by Phillips et al. (2009).
records, Maxent requires a set of pseudo-absence records For model evaluation, 70 % of the species records were
for SDM computation. The selection of these is a crucial used for model training and the remaining 30 % for model
step towards successful model building (Phillips 2008, testing via the Area Under Curve (AUC), referring to the
Mateo et al. 2010, VanDerWal et al. 2009). Herein, we Receiver Operating characteristic Curve (Phillips et al.
randomly selected 20 pseudo-absence records per species 2006, Fielding & Bell 1997). Furthermore, Maxent allows
record within a circular buffer of 10 km as recommended tracing back the relative contribution of each variable to the
by Mateo et al. (2010). This results in a set of pseudo-ab- final model, and so facilitates comparisons with physiologi-
sence records with a similar spatial structure as present in cal properties of the target species. As data splits may affect
39
Correspondence
the model due to the exclusion of some species records that pogenic climate change with A2a and B2a scenarios, which
represent features not present in the remaining records, this forecast nearly identical and only slightly less suitable con-
procedure was repeated 100 times and averages were com- ditions for T. jacksonii (Fig. 4).
puted subsequently. Land cover data for the Hawaii Islands Invasive Hawaiian Jackson’s chameleons are suggested to
were obtained from the Coastal Service Center of the Na- interact with native biota, at least as a predator to endemic
tional Oceanic and Atmospheric Administration (http:// fauna (Holland et al. 2009). While the impact of this spe-
www.csc.noaa.gov/crs/lca/hawaii.html); land cover classes cies is poorly understood, we here provide evidence how
included high- and low-intensity development, cultivated severe the problem may be or may become as T. jacksonii
land, grassland, evergreen forest, scrub/shrub, palustrine finds suitable habitat almost all over the Hawaiian Islands.
emergent wetlands, unconsolidated shoreline, bare land, This emphasizes, in concert with Holland et al. (2009),
and water. Information on current reserve networks was the severity of a problem that should no longer be neglect-
adopted from the Office of Planning, State of Hawaii. ed. Besides the impact on potential prey invertebrates, it
Comparisons of the occupied climate space in T. jackso­ should be considered hypothetically that T. jacksonii might
nii’s native and invaded Hawaiian ranges revealed that the compete with other local fauna for food resources.
highest similarities existed in the ‘precipitation in the wet- The potential invasion success and expected damage
test month’ and the ‘precipitation in the warmest quarter’ to the native fauna may be unparalleled by other invasive
and the greatest differences in the ‘mean monthly tempera- chameleons, although potential distribution and ecological
ture range’, ‘annual temperature range’ and the ‘precipita- impact studies remain limited (e.g. Dimaki et al. 2001). At
tion in the coldest quarter’ (Fig. 3). The ‘minimum tem- least the spatial potential of Chamaeleo africanus in main-
perature of the coldest month’ experienced within T.  jack­ land Greece may be limited as, despite of the species’ oc-
sonii’s invaded range is well above its critical thermal mini- currence on the Peloponesos since historical times, there
mum (5.3 ± 0.5 °C, n = 8, Bennett 2004), but well below is no indication of past or present dispersal (W. Böhme,
its voluntary minimum temperature (29.1 ± 0.32 °C, n = unpubl. data). Likewise, the Réunion population of Furci­
7, Bennett 2004). Its preferred body temperature (32.1 ± fer pardalis is suggested to be restricted to a small, particu-
0.24° C, n = 7, Bennett 2004), voluntary maximum body larly dry and hot area on the western coast near St. Paul,
temperature (34.2 ± 0.31, n = 7, Bennett 2004), panting although this species has recently been found elsewhere on
threshold (36.2 ± 0.67, n =5, Bennett 2004), and critical the island, but its invasion success there remains unknown
thermal maximum (41.0 ± 0.15, n = 8, Bennett 2004; 42.0 (E. Koch, pers. comm., March 2010). Invasion success was
± 0.91° C, n = 10, Lin 1980) are slightly higher than the apparently limited in Furcifer oustaleti (Mocquard, 1894),
maximum temperature of the warmest month within its which was translocated from Madagascar to near Nairobi,
invaded range. Differences of up to 6 °C may be compen- Kenya, but where this species has not been documented
sated by heliothermic thermoregulation, and feeding ac- since 1974 (Spawls et al. 2002). Trioceros jacksonii seems
tivity may occur at temperatures as low as 10 °C (Bennett to be exceptional among chameleons for its invasive poten-
2004), however. tial, as shown by its successful colonisation of the Hawai-
It has been suggested that climatic conditions support- ian Archipelago.
ing successful reproduction and balanced sex ratios can be
strong predictors for the geographic range limits of rep-
tile species (e.g. Rödder et al. 2009b). In T. jacksonii, re-
production is clearly associated with higher precipitation References
during the long wet season in East Africa (Lin 1979, Lin
& Nelson 1981). Although the amount of annual precip- Andrews, R. M. (2008): Lizards in the slow lane: thermal biology
itation is quite similar in the chameleons’ native and in- of chameleons. – Journal of Thermal Biology, 33: 57–61.
vaded ranges, precipitation seasonality is much more pro- Beaumont, L. J., L. Hughes & M. Poulsen (2005): Predicting
nounced in its native range. Thus, the higher monthly pre- species distributions: use of climatic parameters in BIOCLIM
and its impact on predictions of species’ current and future
cipitation in its Hawaiian range may even facilitate circu-
distributions. – Ecological Modelling, 186: 250–269.
mannual breeding.
Following the terminology of Swets (1988), we received Bennett A. F. (2004): Thermoregulation in African chameleons.
– International Congress Series, 1275: 234–241.
‘good’ average AUC values for the SDM-based on current
climate (AUC  = 0.86; AUC  = 0.78). The minimum Böhme, W., A. Bonetti & G. Chiras (1998): The chameleons of
training test the Greek mainland: taxonomic allocation and conservation
prediction value of the training records was 0.12, and the
needs of a second European species. – Herpetozoa, 11: 87–91.
lowest 10 percentile was 0.27. Analysis of variable contri-
butions in the model revealed that on average the ‘precipi- Böhme, W. (2000): When does a foreign species deserve a “per-
mit of residence”? Nonindigenous species (NIS): examples of
tation in the coldest quarter’ had the highest explanative
varying exoticness and varying immigration age, taken from
power (30.2 %), followed by the ‘mean monthly temper-
Herpetology. – Ethology, Ecology & Evolution, 12: 326–328.
ature range’ (26.1 %), and the ‘precipitation seasonality’
Bomford, M., F. Kraus, S. C. Barry & E. Lawrence (2009):
(12.8 %). Average contributions of all other variables were
Predicting establishment success for alien reptiles and am-
< 10 %. Our SDM suggests that, under current ‘bioclimat-
phibians: a role for climate matching. – Biological Invasions,
ic’ conditions, Jackson’s chameleon may find suitable are-
11: 1387–3547.
as nearly everywhere on the Hawaiian Islands, including
Bourgat, R.-M. (1970): Recherches écologiques et biologiques
many reserves, even when the availability of suitable mi-
sur le Chamaeleo pardalis Cuvier, 1829 de l’île de la Réunion
crohabitats in terms of land cover is considered (Fig. 3). et de Madagascar. – Bulletin de la Societé zoologique de la
The situation is not likely to change under future anthro- France, 95: 259–269.
40
Correspondence
Busby, J. R. (1991): BIOCLIM – a bioclimatic analysis and predic- Jaynes, E. T. (1957): Information theory and statistical mechanics.
tion system. – pp. 64–68 in Margules, C. R. & M. P. Austin – Physical Review, 106: 620–630.
(eds.): Nature conservation: cost effective biological surveys Klaver, C. J. J. (1981): Chamaeleo chamaeleon (Linnaeus 1758) –
and data analysis. – Melbourne: CSIRO. Gemeines oder Gewöhnliches Chamäleon. – pp. 218–238 in
Carpenter, A. I., M. Rowcliffe & A. R. Watkinson (2004) The Böhme, W. (ed.): Handbuch der Reptilien und Amphibien
dynamics of the global trade in chameleons. – Biological Con- Europas: Wiesbaden: Akademische Verlagsgesellschaft.
servation, 120: 291–301. Kraus, F. (2009): Alien reptiles and amphibians – a scientific
Dimaki, M. H., E. D. Valakos, A. Legakis, B. Chondropou- compendium and analysis. – Dordrecht: Springer Nether-
los & A. Bonetti (2001): Preliminary results on the feeding lands.
ecology of the African Chamaeleon Chamaeleo africanus Lau- Lin, J. Y. (1979): Fatbody and liver cycles in two tropical lizards,
renti, 1768 from the southwestern Peloponnese, Greece. – pp. Chamaeleo jacksoni and C. hoehneli. – Journal of Herpetology,
57–63 in Lymberakis, P., E. Valakos, P. Pafilis & M. Mylo- 13: 113–117.
nas (eds.): Herpetologia Candiana. – Iraklion: Natural His-
Lin, J. Y. (1980): Desiccation tolerance, and thermal maximum
tory Museum of Crete & Societas Europea Herpetologica.
in two tropical lizards Chamaeleo hoehnelii and C. jacksonii. –
Fielding, A. H. & J. F. Bell (1997): A review of methods for the Copeia, 1980: 363–322.
assessment of prediction errors in conservation presence / ab-
Lin, J. Y. (1981): Comparative reproductive biology of two sym-
sence models. – Environmental Conservation, 24: 38–49.
patric tropical lizards Chamaeleo jacksonii Boulenger and
Flato, G. M. & G. J. Boer (2001): Warming asymmetry in cli- Chamaeleo hoehnelii Steindachner (Sauria: Chamaeleonidae).
mate change simulations. – Geophysical Research Letters, 28: – Amphibia-Reptilia, 3/4: 287–311.
195–198.
Lötters, S., J. Kielgast, J. Bielby, S. Schmidtlein, J. Bosch, M.
Flato, G. M., G. J. Boer, W. G. Lee, N. A. McFarlane, D. Rams- Veith, S. Walker, M. C. Fisher & D. Rödder (2010 “2009”):
den, M. C. Reader & A. J. Weaver (2000): The Canadian The link between rapid enigmatic amphibian decline and the
centre for climate modelling and analysis global coupled mod- globally emerging chytrid fungus. – EcoHealth, 6: 358–372.
el and its climate. – Climate Dynamics, 16: 451–467.
Mateo, R. G., T. B. Croat, A. M. Felicísimo & J. Muñoz (2010):
Giovanelli, J. G. R., C. F. B. Haddad & J. Alexandrino (2008): Profile or group discriminative techniques? Generating reli-
Predicting the potential distribution of the alien invasive able species distribution models using pseudo-absences and
American bullfrog (Lithobates catesbeianus) in Brazil. – Bio- target-group absences from natural history collections. – Di-
logical Invasion, 10: 585–590. versity and Distributions, 16: 84–94.
Gordon, C., C. Cooper, C. A. Senior, H. T. Banks, J. M. Gre- McKeown, S. (1991): Jackson’s chameleons in Hawaii are the re-
gory, T. C. Johns, J. F. B. Mitchell & R. A. Wood (2000): cently described Mt. Kenya subspecies. – Bulletin of the Chi-
The simulation of SST, sea ice extents and Ocean heat trans- cago Herpetological Society, 26: 49.
ports in a version of the Hadley Centre coupled model without
McKeown, S. (1996): A field guide to reptiles and amphibians in
flux adjustments. – Climate Dynamics, 16: 147–168.
the Hawaiian islands. – Diamond Head Publishing Inc, Los
Gordon, H. B. & S. P. O´Farrell (1997): Transient climate Osos.
change in the CSIRO coupled model with dynamic sea ice. – Pearson, R. G., C. J. Raxworthy, M. Nakamura & A. T. Peter-
Monthly Weather Review, 125: 875–907. son (2007): Predicting species distributions from small num-
Hagey, T. J., J. B. Losos & L. J. Harmon (2010) Cruise foraging bers of occurrence records: a test case using cryptic geckos in
of invasive chameleon (Chamaeleo jacksonii xantholophus) in Madagascar. – Journal of Biogeography, 34: 102–117.
Hawai’i. – Breviora, 519: 1–7. Peterson, A. T. & D. A. Vieglais (2001): Predicting species inva-
Heikkinen, R. K., M. Luoto, M. B. Araújo, R. Virkkala, W. sions using ecological niche modeling: new approaches from
Thuiller & M. T. Sykes (2006): Methods and uncertainties bioinformatics attack a pressing problem. – BioScience, 51:
in bioclimatic envelope modeling under climate change. – 363–371.
Progress in Physical Geography, 30: 751–777. Phillips, S. J. (2008): Transferability, sample selection bias and
Hernandez, P. A., C. H. Graham, L. L. Master & D. L. Albert background data in presence-only modelling: a response to
(2006): The effect of sample size and species characteristics on Peterson et al. (2007). – Ecography, 31: 272–278.
performance of different species distribution modeling meth- Phillips, S. J., R. P. Anderson & R. E. Schapire (2006): Maxi-
ods. – Ecography, 29: 773–785. mum entropy modeling of species geographic distributions. –
Hijmans, R. J., S. E. Cameron, J. L. Parra, P. G. Jones & A. Ecological Modelling, 190: 231–259.
Jarvis (2005): Very high resolution interpolated climate sur- Phillips, S. J. & M. Dudík (2008): Modeling of species distribu-
faces for global land areas. – International Journal of Climato- tions with Maxent: new extensions and comprehensive evalu-
lo gy, 25: 1965–1978. ation. – Ecography, 31: 161–175.
Hijmans R. J., L. Guarino & E. Rojas (2002): DIVA-GIS. A geo- Phillips, S. J., M. Dudík, J. Elith, C. H. Graham, A. Lehmann,
graphic information system for the analysis of biodiversity J. Leathwick & S. Ferrier (2009): Sample selection bias and
data. Manual. – International Potato Center, Lima. presence-only distribution models: implications for back-
Holland, B. S., S. L. Montgomery & V. Costello (2009): A ground and pseudo-absence data. – Ecological Applications,
reptilian smoking gun: first record of invasive Jackson’s cha- 19: 181–197.
meleon (Chamaeleo jacksonii) predation on native Hawai- Pope, V., M. L. Gallani, P. R. Rowntree & R. A. Stratton
ian species. – Biodiversity and Conservation: DOI 10.1007/ (2000): The impact of new physical parameterizations in the
s10531–009–9773–5. Hadley Centre climate model: HadAM3. – Climate Dynam-
IPCC (2007): Contribution of Working Group I to the fourth as- ics, 16: 123–146.
sessment report of the Intergovernmental Panel on Climate Ramirez, J. & A. Jarvis (2008): High resolution statistically
Change. – Cambridge, UK, and New York, USA: Cambridge downscaled future climate surfaces. – Bogota: Centre for In-
University Press. ternational Tropical Agriculture.
41
Correspondence
Raxworthy, C. J., E. Martinez-Meyer, N. Horning, R. A.
Nussbaum, G. E. Schneider, M. A. Ortega-Huerta & A. T.
Peterson (2003): Predicting distributions of known and un-
known reptile species in Madagascar. – Nature, 426: 837–841.
Rödder, D. (2009): ‘Sleepless in Hawaii’ – does anthropogenic
climate change enhance ecological and socioeconomic im-
pacts of the alien invasive Eleutherodactylus coqui Thomas
1966 (Anura: Eleutherodactylidae)? – Northwestern Journal
of Zoology, 5: 16–25.
Rödder, D. & S. Lötters (2010): Potential distribution of the
alien invasive Brown tree snake, Boiga irregularis (Reptilia:
Colu bridae), under current and changing climate. – Pacific
Science, 64: 11–22.
Rödder, D. & U. Schulte (2010) Potential loss of genetic vari-
ability despite well established network of reserves: The case
of the Iberian endemic lizard Lacerta schreiberi. –Biodiversity
and Conservation, 19: 2651–2666.
Rödder, D., F. Weinsheimer & S. Lötters (2010a): Molecules
meet macroecology – combining Species Distribution Models
and phylogeographic studies. – Zootaxa, 2426: 54–60.
Rödder, D., J. Kielgast & S. Lötters (2010b): Future potential
distribution of the emerging amphibian chytrid fungus under
anthropogenic climate change. – Diseases of Aquatic Organ-
isms, DOI: 10.3354/dao02197.
Rödder, D., J. Kielgast, J. Bielby, S. Schmidtlein, J. Bosch, T.
W. J. Garner, M. Veith, S. Walker, M. C. Fisher & S. Löt-
ters (2009a): Global amphibian extinction risk assessment
for the panzootic chytrid fungus. – Diversity, 1: 52–66.
Rödder, D., A. Kwet & S. Lötters (2009b): Translating natu-
ral history into geographic space: a macroecological perspec-
tive on the North American Slider, Trachemys scripta (Rep-
tilia, Cryptodira, Emydidae). – Journal of Natural History, 43:
2525–2536.
Rödder, D., S. Schmidtlein, M. Veith & S. Lötters (2009c):
Alien invasive Slider in unpredicted habitat: a matter of niche
shift or of variables studied? – PLoS ONE, 4: e7843.
Rödder, D. & F. Weinsheimer (2009): Will future anthropogen-
ic climate change increase the potential distibution of the alien
invasive Cuban treefrog (Anura: Hylidae)? – Journal of Natu-
ral History, 43: 1207–1217.
Spawls, S., K. M. Howell, R. C. Drewes & J. Ashe (2002): A
field guide to the reptiles of East Africa. – San Diego etc.: Aca-
demic Press.
Swets, K. (1988): Measuring the accuracy of diagnostic systems.
– Science, 240: 1285–1293.
VanDerWal, J., L. P. Shoo, C. H. Graham & S. E. Williams
(2009): Selecting pseudo-absence data for presence-only dis-
tribution modeling: how far should you stay from what you
know? – Ecological Modelling, 220: 589–594.
Waring, G. H. (1997): Preliminary study of the behavior and
ecology of Jackson’s chameleons of Maui, Hawaii. – Report
for USGS/BRD/PIERC Haleakala Field Station presented by
Hawaii Ecosystems at Risk (HEAR) project. Available through
www.hear.org.
Weinsheimer, F., A. A. Mengistu & D. Rödder (2010): Poten-
tial distribution of threatened Leptopelis spp. (Anura, Ar thro-
leptidae) in Ethiopia derived from climate and land-cover
data. – Endangered Species Research, 9: 117–124.
Wisz, M. S., R. J. Hijmans, A. T. Peterson, C. H. Graham, A.
Guisan & N.P.S.D.W. Group (2008): Effects of sample size on
the performance of species distribution models. – Diversity
and Distributions, 14: 763–773.
42